I'm not sure I understand what you mean by visualization, but for example the nitrogen vacancy in InN has its levels in the conduction band. You can look up papers by Chris Van de Walle and Wladek Walukiewicz to get an idea about the behavior and properties of this defect (PRB, APL, PRL).
This case of so called "virtual bonded states". Such defects observed in alloys doped metals with different valence. For exaple Cu doped by Ni or Al doped by Mn. Theory of such defects developed by JacquesFriedel. Diffraction techniques (X-ray, neutrons, electrons) are appropriate for determine of position of such defects in crystal lattice. But it is possible only at high enough concentrations of defects, at least 0,1 mol. %. At low concentrations sometimes may be useful EPR (electron paramagnetic resonance) technique. But direct visualization, I think, possible only with the halp of STM technique. At low concentrations it is very difficult problem.
I never listen about such defect in wide band insulators but for semimetals as TiTe2 and narrow band semiconductors as CdHgTe it is qute ordinary situation.
I meet this phenomenon in intercalation materials based on TiTe, see, for example,
Intercalated metals form d-like bands for transition metals or s-like band for noble and alkali metals near the Fermi level above bottom of conduction band.